Hmm Question maybe someone can help...

[Protege5Fizzle]

The protege' runs on a preset map when the ecu sees WOT correct?
If so what if you rewire the throttle position sensor so it dosent see wide open throttle and dosent dump an assload of fuel? would this be benifit the engine in making power at WOT for us N\A peeps by causing it to run not as rich? or is this sounding rediculous?
Only reason I ask Is becuase I've noticed the car pulls harder when you dont have the throttle all the way open, and also it runs closer to stoich. But when you press that pedal to the floor it seems to take a dump and runs very rich. But say try and trick the ecu by not letting it see WOT to lean it out a little. Maybe this has already been done\attempted and I cant find it anywhere but lemme know guys.

[Protege5Fizzle]

Oh and if it helps I believe the ecu sees between 3.0-4.6V at WOT and .1-1.1V at CTP so maybe changing the signal wire to never send anything over 2.5-2.9 back to the ecu... Maybe? or should I just shoot myself in the foot for thinking too much about this?

[Super Matty P]

I guess I'll tell you guys here but just keep it here. I dont want this on other forums.

I spent some time with a friend who is an electrical engineer. He and I developed a piece similar to an exhaust MIL that splices onto your stock TPS and will actually peel off voltage up to WOT.

Let me elaborate.

we have a 0-5V TPS signal. As the butterfly opens the tps sees an increase in voltage up to WOT (5v)... what my pack does is start making small reductions in the output voltage of the tps so that when you get to WOT (5v) you're only at 4.75v. The throttlebody will still be open and the engine will be receiving the air it would need it's just that the ecu would never see WOT.

I don't know for sure. I've never tested it but the engine may go into closed loop under a load...not just WOT. If thats the case then this would do nothing. If it DOES work then piggyback controllers like the S-AFCII are now available to us.

I'll post a pic of it....anyone wanna try it?

[Super Matty P]

the connections are simple.... tps signal in, signal out and tps (or chasis) ground. The signal wire provides it's own power.

[SilentSno]

Theory all sounds right....

Matty that it a hot piece right there. :)

[Peepsalot]

Well, I'm not exactly sure how the mazda ECU uses the TPS signal, but I think mostly the fuel delivery is determined by the MAF & IAT signals.
The TPS sensor tells MegaSquirt what the current position of the throttle. This variable is compared to the most recent readings to determine if the throttle is opening or closing rapidly. If so, extra fuel can be added for an opening throttle to compensate for transient conditions. This functions the same as an accelerator pump in a carburetor.

The TPS also does two other important functions. First, if the throttle is open more than a specified amount during cranking, it invokes the “flood clear” mode by reducing the injected pulse width to 0.3 milliseconds. Second, if the throttle is open more than 70%, exhaust gas feedback is shut off.

That is what megasquirt uses the TPS for, so if the mazda ECU works in any similar way, then altering tps only affects fuel during quick throttle variations.

[t3ase]

I guess I'll tell you guys here but just keep it here. I dont want this on other forums.

I spent some time with a friend who is an electrical engineer. He and I developed a piece similar to an exhaust MIL that splices onto your stock TPS and will actually peel off voltage up to WOT.

Let me elaborate.

we have a 0-5V TPS signal. As the butterfly opens the tps sees an increase in voltage up to WOT (5v)... what my pack does is start making small reductions in the output voltage of the tps so that when you get to WOT (5v) you're only at 4.75v. The throttlebody will still be open and the engine will be receiving the air it would need it's just that the ecu would never see WOT.

I don't know for sure. I've never tested it but the engine may go into closed loop under a load...not just WOT. If thats the case then this would do nothing. If it DOES work then piggyback controllers like the S-AFCII are now available to us.

I'll post a pic of it....anyone wanna try it?
Nice. I'm getting a motor put in or reworked, I'm down for trying it. Let's just keep it a FlyinProtege thing, though. Unless it works then you can sell them for $400. ;D

[Protege5Fizzle]

hmm very nice matty, hopefully it'll work... Dont see why it wouldnt but then again knowing mazda. But anywho I say we test this thingy out. If ya have a diagram or anything on how its made Lemme know. I'll make one and test it er something.

[Super Matty P]

What I wanted to do was find someone willing to install an S-AFCII on their car. They'd go and get some baseline dyno pulls, see what needs to be adjusted, then install the tps pack, make the changes to the s-afc and then re-dyno (all same day for consistancy).

This will be the ultimate test since everyone says the ecu will throw out the piggybacks at WOT.

Another thing is that I may not have sheered off enough voltage at WOT. Maybe the stock ecu sees WOT at 4.5 or higher and I've got it set to limit voltage to 4.7.

If it works it'll be nice and I can make as many as people are willing to buy.

[Gen1GT]

I'd just like to add as a sidenote though, our cars(all gens) are never actually fully open loop, even at WOT. Parameters change, so the ECU has to adjust, even at WOT. ie, during a -20°C cold start in the Rockies at WOT, the car will use different fuel and ignition maps than during a 40°C, humid day at sea level at WOT.

[Peepsalot]

I'd just like to add as a sidenote though, our cars(all gens) are never actually fully open loop, even at WOT. Parameters change, so the ECU had to adjust, even at WOT. ie, during a -20°C cold start in the Rockies at WOT, the car will use different fuel and ignition maps than during a 40°C, humid day at sea level at WOT.
If the ECU is not using feedback from the O2 sensor to keep the AFR right at 14.7, then it's in open loop. I know it still looks at a variety of sensors to determine the amount of fuel to use, but it's still considered open loop. A closed loop system involves taking the results of previous output, and using that to compute the next output.

[Super Matty P]

is anyone interested in doing the S-aFCII thing? If so I'll give them the tps pack if they'll B&A dyno.

[Gen1GT]

If the ECU is not using feedback from the O2 sensor to keep the AFR right at 14.7, then it's in open loop. I know it still looks at a variety of sensors to determine the amount of fuel to use, but it's still considered open loop. A closed loop system involves taking the results of previous output, and using that to compute the next output.

An open loop control is not an actual circuit. An open loop control system is controlled by a single input signal. And example would be an IRTB setup, which would control A/F ratios solely on throttle position input.

Fuel injectors and coil packs are each a type of open loop control. Each with a single input and output. Input comes from the ECU of course. Adding variables with other sensors, such as O2 feedback, temp sensor, MAP, MAF or other, makes it a closed loop control circuit, with many variables.

To my understanding OBDII cars can run fully open loop during safe modes, using only TPS. Don't quote me on that though.

[BRIAN MP5T]

Yeah, What's the deal with the A/F Holding at loads over like 130 Km/h. I know it's supposed to save the motor, is there a way around that?

[t3ase]

Matty, we need to talkz0r.

[Super Matty P]

ok, just shoot me a pm.

[Mr.Win]

Interesting turn of events... man i love this place.

I truly hope this works!

[t3ase]

Well, he's got a local guy that may be helping him. If that falls through, I'll be next up to try this thing out.

Seems after that talk we had that it looks like it may work nicely. We'll see.

[Super Matty P]

the local guy still wants to do it but he doesn't have the money for the s-afc right now.

I dont think just wiring this tps pack up will produce any gain so without an AFC of some kind trials are pointless.

[t3ase]

the local guy still wants to do it but he doesn't have the money for the s-afc right now.

I dont think just wiring this tps pack up will produce any gain so without an AFC of some kind trials are pointless.
It's worth a dyno run anyway, before he "enables" AFC control. Do three, stock, with mod, with mod+afc.

[Super Matty P]

what I was thinking was doing 4-6 pulls....the first half for baselines...then use those averages to set your adjustment....THEN dyno the last 2-3.

[t3ase]

what I was thinking was doing 4-6 pulls....the first half for baselines...then use those averages to set your adjustment....THEN dyno the last 2-3.
Heatsoak is going to kill me with my setup, unless I manage to slip a FMIC in this rebuild.

[Super Matty P]

you should be fine with the hood up and the huge dyno fan up against the grill

[KzA]

very interesting..great thinking guys..

The only reason the AFC doesnt work is becuase of the map change at WOT? Is this the same for other piggybacks that aren't available to us? IF so, this could open up a slew of new options..

[Super Matty P]

that was my thoughts. I don't know if the loop switch is based on engine load or throttle position. It could be either/or/both. If it's just based on TPS then this should fix it. The only reason why it wouldn't is if I didn't sheer enough voltage off. That's an easy fix though.

[Peepsalot]

Matty, what is it about the way the SAFC-II works that makes you think this will help? I just re-read this thread, I misunderstood what you guys were trying to do at first. Sounds like you want the car to stay out of open loop mode? I don't see how that helps with tuning though. Usually you want to do the opposite, and keep it in open loop.

[Peepsalot]

This did give me a pretty cool idea for a tuning device that I don't think has been done before. Basically, it would work in conjunction with a wideband and it would replace the ECU's O2 signal with a modified one, so that you could set target AFR's to anything you want instead of the normal 14.7.

[Super Matty P]

The way I understand the ECu operation is that at WOT the ecu shuts out sensor input and will operate on preset maps. The S-afc is supposedly proven to work for cruising but at WOT its settings get overridden. This should keep that override from happening and allowing you the ability to tune your AFRs no matter how you want to drive it. This would be a gain for NA or FI proteges. I think thats what most people are looking for.

[t3ase]

The way I understand the ECu operation is that at WOT the ecu shuts out sensor input and will operate on preset maps. The S-afc is supposedly proven to work for cruising but at WOT its settings get overridden. This should keep that override from happening and allowing you the ability to tune your AFRs no matter how you want to drive it. This would be a gain for NA or FI proteges. I think thats what most people are looking for.
Correct.

[t3ase]

QUestion, though does this "device" pull voltage in a linear fashion? Meaning, if sensor is reading 5v, it brings it to 4.75 but if it's at 1v, will it pull to .75?

[KzA]

by the way he described it before, it sounded like the voltage stays the same up until it becomes near WOT, then it slowly takes away voltage...so I actually think it is linear but not until a certain voltage..but I'll wait until he chimes in to let us know for sure

[Super Matty P]

No, what it does is very slowly pulls voltage away the entire time.

this is just example numbers..

@ 1V the ecu will see .99
@ 2V the ecu will see 1.95
@ 3V the ecu will see 2.90
@ 4V the ecu will see 3.85
@ 5V the ecu will see 4.80

Thats almost exactly how it's set up but the final viltage output should be just a hair over 4.75V. I wanted to keep as much voltage as possible while still taking off enough to keep it off WOT.

[t3ase]

So it will be progressive.

[KzA]

interesting..

[Peepsalot]

No, what it does is very slowly pulls voltage away the entire time.

this is just example numbers..

@ 1V the ecu will see .99
@ 2V the ecu will see 1.95
@ 3V the ecu will see 2.90
@ 4V the ecu will see 3.85
@ 5V the ecu will see 4.80

Thats almost exactly how it's set up but the final viltage output should be just a hair over 4.75V. I wanted to keep as much voltage as possible while still taking off enough to keep it off WOT.Lemme guess, your device consists of two resistors. :p

[KzA]

http://www.flyinprotege.com/forums/attachment.php?attachmentid=134&d=1138955474

[Super Matty P]

sorta....it's a little more involved than that but it's along those lines.

[SpicyMchaggis]

buy me one, i'll do it.

[Dexter]

Lemme guess, your device consists of two resistors. :p

voltage divider yay